The effects of many single stressors have been reported,but how pigs perform when subjected to more than one ortwo stressors at a time, as is common in commercialswine production, has not. To study this, 256 YorkshireH Hampshire or purebred Duroc pigs (34.7 " 0.5 kg) weresubjected to one of the eight treatment combinations (2H 2 H 2 factorial) of ambient temperature [consantthermoneutral (24° C) or high cycling temperature(28-34° C)], stocking density (.56 m² or .25m²/pig) and social group (static group or regroupedat the start of wk 1 and 3) during a 4-wk experiment.The stress of high temperature, high stocking densityand regrouping depressed 4-wk ADG by 12%, 16% and 10%(P<.05); and ADFI by 7%, 6% and 5%, respectively(P<.05). Out of a possible 60 stressor interactionsfor ADG, ADFI and G:F, there were no significantthree-way interactions and only six two-wayinteractions, suggesting the effects of the individualstressors were additive (i.e., the stressors' effects onperformance were similar regardless of whether they wereimposed singly or in combination). Thus, when pigs weresubjected to all three stressors simultaneously ADG,ADFI, and G:F were depressed by 31%, 15%, and 18%,respectively. Stressor additivity was furthercorroborated by examining the effect of "stressororder," or the number of stressors imposedsimultaneously. As the number of stressors increasedfrom 0 to 3, ADG, ADFI and G:F decreased linearly. Thesedata suggest that multiple concurrent stressors affectgrowth performance of pigs in a predictable fashion(i.e., additively) and indicate that avoidance orremoval of a given stressor is advantagous even whenother uncontrollable stressors persist.

Introduction

The effects of many single stressors and a few twostressor complexes have been reported for swine. Whilethe results indicate that stressors such as high ambienttemperature (Close et al., 1978; Lopez et al., 1991;Nienaber et al., 1991; Xin and DeShazer, 1992),regrouping (Bjork et al., 1988; McGlone and Curtis,1985), and restricted floor space (Kornegay and Notter,1984; Kornegay et al., 1993a,b; NCR-89 Committee onConfinement Management of Swine, 1993), reduce feedintake and weight gain, it is difficult to extendinformation from single stressor studies to productionsettings because pigs usually experience severalstressors at once. Recently, the effects of sixconcurrent stressors and effects of sequential stressorson performance and several physiological andpathological traits in the chick have been reported(McFarlane et al., 1989a,b,c; Johnson et al., 1991).When factorial combinations of six stressors wereimposed, effects generally were additive. For example,the percentage depression in feed intake and growth,respectively, increased linearly as the number ofsimultaneously imposed stressors increased. Whether theeffects of multiple concurrent stressors in swine areadditive, synergistic or antagonistic is not known. Abetter understanding of how several stressors actingtogether affect pigs' overall well-being and performancewill enhance progress in management of animalenvironments.

Materials and Methods

Two hundred and fifty-six pigs (Yorkshire and Hampshirecrossbreds and purebred Durocs) with an initial bodyweight of 34.7 ".50 kg were used in two 4-wk trials.Within each trial, after taking into consideration bodyweight, gender and litter of origin, pigs were assignedto one of eight treatments in a 2³ factorialarrangement. Treatments were imposed in two adjacentmechanically ventilated rooms. Two environmentaltemperatures (thermoneutral 24° C and hot diurnaltemperature cycling from 28 to 34° C) were imposedin each room across two trials. The 2² factorialarrangement of space allocation (.56 and .25m²/pig) and regrouping (static group or regrouped)were imposed on a pen-within-room basis.

The two rooms were virtually identical, each havingeight pens that were each equipped with a two-holefeeder, nipple waterer and a partially slotted floor.Both rooms were mechanically ventilated and hadadjustable baffle inlets along one side that wereadjusted to provide uniform air flow and velocity inboth rooms. The air exchange rate was adjusted to be thesame in both rooms to minimize differences in airquality. Aerial ammonia and hydrogen sulfideconcentrations were measured at pig level at the centerof each room during week 1 and 3 to ensure relativelyequal air quality. Ammonia concentration in thethermoneutral and high ambient temperature roomsaveraged 5 and 4 ppm, respectively. Hydrogen sulfide wasnot detected in either room. Relative humidity andtemperature were monitored at pig level throughout thestudy using a hygrothermograph. A 24-h lighting regimenwas employed in both rooms.

Pigs were provided ad libitum access to acorn-soybean meal-based diet that was formulated toexceed NCR (1988) nutrient requirements for grower pigs(17% CP, .9% lysine, and 3296 Kcal ME/kg). Eight pigswere assigned to each pen, and following a 1-wkacclimation period at .56 m²/pig and 24° C,stressors were imposed. The room used for the hotdiurnal environment was programmed to cycle from a lowof 28° C from 2400 h to 0600 h to a high of 34°C from 0700 h to 2300 h. Pen size was adjusted toprovide .25 m²/pig for pigs assigned a treatmentrequiring restricted space allocation. Social stress wasinduced at the beginning of week 1 and 3 by regroupingpigs. This was accomplished as follows: Within each roomfor each experiment the 2² factorial arrangement ofspace allocation and regrouping was replicated twice.Four randomly selected pigs from one treatment replicatewere switched with four randomly selected pigs from theother (i.e., pigs switched pens but not treatments). Atthe beginning of week 3, four pigs from one replicatewere regrouped with the four unfamiliar pigs from theother, creating again, two treatment replicates withunfamiliar pigs.

Pig body weight and feed disappearance were determinedat 1-wk intervals during the 4-wk experiment so thatADG, ADFI, and G:F could be calculated. Each pen wasvideo-recorded for 24 h during week 2 and 4. Theduration of standing, lying, and sitting behavior wasestimated by scan sampling every 15 min. At the end ofeach trial, blood samples were collected from 4 randomlyselected pigs from each pen. Blood smears were made induplicate from each sample and stained for differentialleukocyte counts. Basal plasma cortisol was determinedas was the cortisol response of pigs to exogenous ACTHafter treatment with dexamethasone.

All data were analyzed with the PROC GLM procedures ofSAS (1990). The statistical model included the maineffects of trial, temperature, space allowance andregrouping, and all two- and three-way stressorinteractions. Data were also categorized according tothe number of stressors making up the stressor complex.Pigs were subjected to 0 (control), 1 (either highdiurnal temperature, regrouping, or reduced spaceallocation), 2 (high diurnal temperature and regrouping;high diurnal temperature and reduced space allocation;and regrouping and reduced space allocation), or 3stressors (high diurnal temperature and regrouping andreduced space allocation). Data were subjected tocurvilinear regression analysis to detect whether therewas deviation from a linear response (McFarlane et al.,1989).

Results

To study the effects of multiple concurrent stressors ongrowth performance, pigs were subjected to one of theeight treatment combinations (2 H 2 H 2) ofenvironmental temperature (24° C or 28-34° C),space allowance (.56 m² or .25 m²/pig), andsocial group (static group or regrouped) in a 4-wkexperiment. The main effects of each stressor for ADG,ADFI and G:F for the entire trial and during weeks 1, 2,3 and 4 are shown in Table 1. Exposure to high cyclingtemperature, reduced space allowance, or regrouping,depressed ADG over the 4-wk experiment by 11.9%, 16.4%,and 9.6%, respectively (P<.05). Stressors alsodepressed feed intake: Average daily feed intake forpigs subjected to high cycling temperature, reducedspace allowance, or regrouping was depressed by 7.4%,6.0%, and 5.0%, respectively (P<.05; Table 1).Whereas temperature and regrouping did not influence4-wk G:F (P>.05), reduced space allowance depressedfeed conversion efficiency by more than 10%. Thedepression in growth, feed intake and feed efficiency bystressors varied from one week to the next. Forinstance, reduced space allowance and high cyclingtemperature depressed ADG and ADFI more during week 4than during week 1 (Table 1). In fact, growthperformance was not affected by temperature until week 3and 4.

Of the possible 60 interactions, only 6 weresignificant. A temperature H regrouping interaction wasdetected for ADG during week 1, 3 and 4, and for G:Fduring week 1 (P<.05). During week 1 and 3, theeffect of regrouping was greater in the 24° Cthermoneutral environment than in the high cyclingtemperature. The opposite was true during week 4,however. A space allowance H regrouping interaction wasdetected for ADG during week 2 and over the entire 4-wkexperiment (P<.05). The interaction indicated thatthe effect of regrouping was less when space allowancewas restricted.

The 4-wk ADG, ADFI and G:F for pigs on each of the eighttreatment combinations are presented in Table 2. Thedepression of ADG, ADFI, and G:F caused by imposing eachstressor individually was similar (P>.05). Inaddition, there were no differences in the ADG, ADFI,and G:F for pigs subjected to the different two-stressorcomplexes. Overall, when pigs were subjected to thethree stressors simultaneously, ADG, ADFI, and G:F weredepressed by 30.8%, 15.1%, and 17.5%, respectively. Thetreatments were also categorized according to the numberof stressors they comprised (see Table 2). Regressionanalysis detected a linear and negative effect of thenumber of stressors imposed on ADG, ADFI, and G:F(Figure 1).

No behavioral or physiological parameter examined provedto be particularly useful for predicting the effects ofstress on growth performance of pigs with the possibleexception of the neutrophil:lymphocyte ratio. Regressionanalysis detected a linear and positive effect of thenumber of stressors imposed on the neutrophil:lymphocyteratio, but the model explained only 7% of the treatmentvariation. The packed cell volume, basal plasma cortisolconcentration, and plasma cortisol after injection ofdexamethasone and ACTH were not different between pigssubjected to the different stressors.

Discussion

To understand how different stressors that are presentsimultaneously influence growth performance, pigs weresubjected to a factorial arrangement of treatmentsinvolving three stressors. The resultant data confirmprevious studies which show that when presented alone,high ambient temperature, reduced space allowance, andregrouping depress growth, feed intake, and feedconversion efficiency. The important finding was thatwhen these stressors were presented together, as isoften the case in typical swine production systems, theeffects were in general, additive. Because the effectswere additive, these data indicate that removal of asingle stressor can have substantial beneficial effectson the growth performance of pigs, even when a number ofother uncontrollable stressors persist. They furthersuggest that it may be possible to predict the effectsof multiple simultaneous stressors on growth performanceof pigs if the effects of the individual stressors areknown.

Because the underlying goal of this study was todetermine if the effects of multiple stressors on growthperformance of pigs were additive, synergistic orantagonistic, it was important that each individualstressor induce a significant depression in growthperformance. Therefore, the type of stressor and thelevel of stress were chosen judiciously in order toensure a detrimental effect on performance, but to stillbe within limits of what may be observed in typicalproduction systems. Accordingly, the main effects ofhigh ambient temperature, reduced space allowance andregrouping significantly depressed ADG and ADFI. Thesefindings were predictable and are in agreement with anumber of previous studies that reported the effects ofsimilar stressors on growth performance of pigs.

That the three stressors were found to have significantnegative effects on growth performance suggested that byimposing them simultaneously in the various factorialcombinations one could effectively evaluate howdifferent stressors interacted. Surprisingly, out of apossible 60 stressor interactions for ADG, ADFI and G:F,there were no three-way interactions and only 6 two-wayinteractions. In most cases, whereas in one week theinteraction was antagonistic, in another week it wassynergistic. The net effect was that for the performanceover the 4-wk experiment, only one significantinteraction was detected. A space allowanceH regrouping interaction was detected forADG indicating the effect of regrouping was less whenspace allowance was restricted. That reduced spaceallowance increases the frequency of challenges to apigs' personal space and, therefore, results in anincrease in agonistic encounters, regrouping pigs thathad reduced space allowance may not have been asstressful (i.e., perhaps pigs kept under reduced spaceallowance were acclimated to a relatively high level ofagonistic behavior). Thus, if reduced space allowanceand regrouping impinged upon the pig via a similar path(i.e., they both caused psychological stress andincreased agonistic behavior), this may explain theantagonistic interaction. It should be noted, however,that other stressors which impinge upon the pig via acommon path may exacerbate one another as is the case ofaerial ammonia reducing the ability of pigs to clearbacteria from their lungs (Drummond et al., 1978) andincreasing the nasal-turbinate shrinkage in pigsinfected with Bordetella bronchiseptica (Drummondet al., 1981). Nonetheless, in the present study theeffects of stressors were by-and-large,additive.

To further examine the additivity of stressors,treatments were categorized according to the number ofstressors they comprised. This ranged from 0 for controlto 3 for the treatment comprising all of the stressors.These data were subjected to curvilinear regressionanalysis to determine if increasing the number ofstressors from 0 to 3 decreased performance in a linearor quadratic fashion. McFarlane et al. (1989b) used thisapproach to study the effects of six concurrentstressors on chick performance. They found that as thenumber of stressors increased, performance decreasedlinearly. When chicks were subjected to six stressorssimultaneously, their growth rate over a 7-d trial wasdecreased by more than 60%. These authors also reportedvery few stressor interactions. The results of thecurrent study are remarkably consistent with the resultsreported by McFarlane et al. (1989b) for chicks. Forinstance, there was a linear and negative relationshipbetween the number of stressors imposed and ADG, ADFIand G:F. When pigs were subjected to three stressorssimultaneously, ADG, ADFI, and G:F were depressed by30.8%, 15.1%, and 17.5%, respectively. The regressionequation for ADG indicates that the addition of eachstressor resulted in an additional 10% depression ingrowth. It is important to point out, however, that notevery stressor contributes equally as this may suggest.Moreover, it is not the mere presence or absence of astressor that is important, but also the level of stressinduced. For instance, pigs of this size may experienceheat stress if kept at either 34° C or 30° C,but clearly the depression in growth would be greater at34° C. Nonetheless, the present data suggest that insome cases the effects of multiple concurrent stressorson growth performance of pigs is equal to the sum of thestressors' individual effects.

Implications

Animal environments are complex and frequently compriseseveral stressors simultaneously. Despite this, whetherstressor effects on growth performance are additive,antagonistic or synergistic is not known. Thisinformation is needed to better manage animals'environments. These data show that the detrimentaleffects of high ambient temperature, regrouping andreduced space allowance are, in general, additive.Therefore, removing a single stressor from a complexenvironment containing multiple stressors maysubstantially improve growth performance of pigs. Thisis important because some stressors are uncontrollableand therefore may be present even under optimalmanagement. Furthermore, that the multiple concurrentstressors affected performance in a predictable manner(i.e., additively), it may be possible to estimate thegrowth response of pigs in a complex environmentcomprising multiple stressors if the effects of theindividual stressors are known.

References

Björk, A., N. G. Olsson, E. Christensson, K. Martinssonand O. Olsson. 1988. Effects of amperozide on bitingbehavior and performance in restricted-fed pigs followingregrouping. J. Anim. Sci. 66:669.

Close, W. H., L. E. Mount and D. Brown. 1978. Theeffects of plane of nutrition and environmentaltemperature on the energy metabolism of the growing pig.2. Growth rate, including protein and fat absorption.Br. J. Nutr. 40:423.

Drummond, J. G., S. E. Curtis, R. C. Meyer, J. Simon andH. W. Norton. 1981. Effects of atmospheric ammonia onyoung pigs experimentally infected with Bordetellabronchiseptica. Am. J. Vet. Res. 42:963.

Drummond, J. G., S. E. Curtis and J. Simon. 1978.Effects of atmospheric ammonia on pulmonary bacterialclearance in the young pig. Amer. J. Vet. Res.39:211.

Johnson, R. W., S. E. Curtis and R. D. Shanks. 1991.Effects on chick performance of ammonia and heatstressors in various combination sequences. Poultry Sci.70:1132.

Kornegay, E. T., M. D. Lindemann and V. Ravindran.1993a. Effects of dietary lysine levels on performanceand immune response of weanling pigs housed at two floorspace allowances. J. Anim. Sci. 71:552.

Kornegay, E. T., J. B. Meldrum, and W. R. Chickering.1993b. Influence of floor space allowance and dietaryselenium and zinc on growth performance, clinicalpathology measurements and liver enzymes, and adrenalweights of weanling pigs. J. Anim. Sci. 71:3185.

Kornegay, E. T., and Notter. 1984. Effects of floorspace and number of pigs per pen on performance. PigNews and Information. 5:23.

Lopez, J., R. D. Goodband, G. L. Allee, G. W. Jesse, J.L. Nelssen, M. D. Tokach, D. Spiers and B. A. Becker.1994. The effects of diets formulated on an idealprotein basis on growth performance, carcasscharacteristics and thermal balance of finishing giltshoused in a hot, diurnal environment. J. Anim. Sci.72:367.

McFarlane, J. M. and S. E. Curtis. 1989a. Multipleconcurrent stressors in chicks. 3. Effects on plasmacorticosterone and the heterophil:lymphocyte ratio.Poultry Sci. 68:522.

McFarlane, J. M., S. E. Curtis, R. D. Shanks and S. G.Carmer. 1989b. Multiple concurrent stressors in chicks.1. Effect on weight gain, feed intake, and behavior.Poultry Sci. 68:501.

McFarlane, J. M., S. E. Curtis, J. Simon and O. A.Izquierdo. 1989c. Multiple concurrent stressors inchicks. 2. Effects on hematologic, body composition, andpathologic traits. Poultry Sci. 68:510.

McGlone, J. J. and S. E. Curtis. 1985. Behavior andperformance of weanling pigs in pens equipped with hideareas. J. Anim. Sci. 60:20.

NRC. 1988. Nutrient requirement of Swine. NationalAcademy Press, Washington, D.C.

NCR-89 Committee on Confinement Management of Swine.1993. Space requirements of barrows and gilts pennedtogether from 54 to 113 kilograms. J. Anim. Sci.71:1088.

Nienaber, J. A., T. P. McDonald, G. L. Hahn and Y. R.Chen. 1991. Group feeding behavior of swine.Transactions of the ASAE. 34:289.

SAS, 1990. SAS User=s Guide: Statistics.Version 6 Edition. SAS Institute, Cary, NC.

Xin, H. and J. A. DeShazer. 1992. Feeding patterns ofgrowing pigs at warm constant and cycle temperatures.Transactions of the ASAE 35:319.